CIMOLIORNIS DIOMEDEUS. 547 
ous wall is thin and compact, and the cavity large and 
smooth, like that of the air-bones in birds of flight. It 
differs from the femur of any known bird in the pro- 
portion of its length as compared with its breadth, and 
from the tibia or metatarsal bone in its trihedral figure 
and the flatness of the sides, none of which are lon- 
gitudinally grooved. It resembles most the humerus of 
the Albatross, both in its form, proportions, and_ size, 
but differs therefrom in the more marked angles which 
bound the three sides. The extremity becomes compressed 
and expanded, like the distal end of the humerus of the 
Albatross, but is too much mutilated to allow of the 
precise degree of similarity or difference to be determined. 
On the supposition that this fragment of bone is the shaft 
of the humerus, its Jength and comparative straightness 
would prove it to have belonged to one of the longipennate 
natatorial birds, equalling in size the Albatross. 
The trihedral form of the shaft of the bone resembles 
that of the upper or proximal half of the ulna of the Al- 
batross ; but there are no distinct traces of the attachments 
of the quill-feathers. By the same trihedral form it may 
be compared with the distal portion of the radius of the Al- 
batross ; but this idea can only be entertained by supposing 
the fossil bird to have been of gigantic dimensions, almost 
realizing the fabulous ‘Roc’ of Arabian romance; and 
the other portions of bone associated with it, and most 
probably parts of the same bird, render this last suppo- 
sition still less probable. 
The most characteristic of these portions (fig. 250) ap- 
pears to be the distal end of the tibia, the peculiar troch- 
lear extremity of which, characteristic of the class of 
Birds, is sufficiently preserved, although crushed. Their 
relative size to the preceding bone, on the supposition 
NWN 2 
